Rempa

Add up to 20 products to a project's product tracking (products add-on), by their page address or their name. This BILLS the account: each active product costs per month $5 (updated every week), $8 (every 3 days) or $12 (every day), 20% less with yearly billing, invoiced now on the subscription, prorated. Confirm the products and the frequency with the person before calling it. Products already tracked are skipped. Rempa then reads each page and drafts its prompts; the first results come within the hour. Refused on the free trial, over the account's product limit, or when the payment fails (then nothing is added). Returns the products added and the new monthly total. An OAuth connection needs the permission to update.

Compare a tracked product head-to-head with one competing product AI names, over 7, 30 or 90 days: in the answers to questions that name neither, how often AI names both, only one or neither; each one's visibility, average position, how often each is named first when both are, sentiment, the typical price AI gives (median, as AI writes it) and the words about price; the same visibility per AI model and per update; and what AI says about each. Default competitor: the main one the brand picked, else the one named most. AI prices are never compared with the real price. Read-only.

Get one AI answer in full: the prompt, the model and its version, the country, the complete answer text (markdown), the brands it names with their position, sentiment (0 to 100) and the reason for it, and its sources (URL, website, title). Use it to read what an AI assistant actually says after spotting an answer with list_answers. Read-only.

Get how much people search for what each prompt asks, to see which prompts matter most: per prompt, the monthly Google searches of equivalent queries in the project's market (a proxy, since AI assistants publish no prompt volumes), the trend over a year, the brand's visibility and best position on it in the latest update, the competitors named, and the opportunity (searches going to answers that do not name the brand). Per topic, the total demand and the share of AI requests about it (Anthropic Economic Index, Claude.ai usage). Read-only.

Get the trend over time: per update date, the visibility, average position, sentiment (0 to 100, 50 neutral) and share of voice of the brand and of up to 6 main competitors. Use it for "are we progressing?" or to see whether a change paid off; for one update in detail, use get_overview. Read-only.

Get the headline numbers of one update, as on the app's dashboard: the brand's visibility (share of answers naming it), average position (1 = named first), share of voice and how often its website is cited; the same per AI model; every brand named with its visibility, position and share of voice; and the 15 websites cited most. The best first call to answer "how visible are we in AI answers?". Defaults to the latest update, all models and all topics. Read-only.

Get how the AI answers of one update perceive the brand and its competitors: per brand, the average sentiment (0 to 100, 50 neutral), the number of positive, neutral and negative answers, and the share of voice (plain, and weighted by position: a mention weighs 1 when named first, 1/2 when second…); plus, for one brand, a summary of its strengths, weaknesses and the words used about it. Use it for "what do AI assistants say about us?". Read-only. The summary is written by AI from the answers the first time it is asked for an update, then kept; requests with a summary are limited to 60 per hour per account, beyond which it comes back as unavailable.

Get one tracked product in detail: its numbers from the latest update, its prompts (discovery ones do not name it and measure its visibility; evaluation ones name it), the answers of its latest update (per prompt and AI model: whether it is named, its position and sentiment, the price the answer gives and the products named), and the competing products named over the period with how often they are named instead of it or ahead of it. Prices are the ones AI gives, as written: never compare them with the real price, which promotions and variants change. Read-only.

Get a project's setup: the brand (name, other names, website, market, profile), the competitors tracked, the AI models asked (their ids are the values for the model filter of the other tools) and the topics of the latest update (values for the topic filter). Use it to know what a project tracks; for results, use get_overview. Read-only.

Get one recommendation with everything needed to carry it out: its steps and evidence (cited pages with their author, the brands they name and their contact page; the prompts and Google queries concerned; the brand's own page to rework), the page published, and its full workspace (the outreach message, the content brief, and the draft's Markdown when one was written in the app). Use it before writing the e-mail or page with the user; never invent facts, reviews or testimonials. Read-only. Paid plans only.

Get the project's action plan to be named more often by AI assistants, built from its latest update and ranked by the monthly searches at stake: websites to get listed on (cited when competitors are named and the brand is not), pages to publish (questions the brand loses), and fixes to its website (AI crawlers blocked in robots.txt, llms.txt, structured data). Each item has an id, status (todo, doing, done, dismissed), summary, steps, impact, effort and evidence; the page published once done (and whether Rempa found it online and naming the brand); the date it is set aside until; and its workspace: the outreach message, content brief (outline, questions, competitors, sources to cite, the brand's pages that should link to it) and draft written in the app, the draft as a title and length (get_recommendation has its text). Use it for "what should we do?"; record progress with update_recommendation. Read-only. Paid plans only: on a free plan it returns an error.

List the websites the AI answers of one update cite, most cited first: per website, the number and share of answers citing it, its citation count and whether it is the brand's own site. Use it to find the review sites, directories and media to be listed on; to see which answers cite one website, use list_answers with its website filter. Read-only.

List the AI answers of one update, one row per prompt and AI model: the prompt, its topic, the model, whether the brand is named and at which position, whether its website is cited, the brands named and the source URLs. Filter by model, topic, a brand named or a website cited, for example to find the prompts where a competitor is named and the brand is not. Paginated: total and next_offset tell whether more remain. Rows hold no answer text; use get_answer with an answer_id for it. Read-only.

List the products a project tracks (products add-on), each asked on its own buyer questions: per product, its id, name, category, page, how often it is updated, whether it is active or paused, and from its latest update its visibility (share of the answers to questions that do not name it where AI names it), average position among the products named, sentiment (0 to 100) and the competing product named most (or the main competitor the brand picked). Also the account's product billing. Use it for "how are our products doing in AI answers?". Read-only.

List the projects of the connected Rempa account. Call this first: every other tool takes a project id or address from here. Returns the account's plan and, per project, its id, address, name, tracked brand, website, whether setup is complete, the date of its latest update and its link in the app. Read-only; takes no input.

List a project's updates (runs), newest first, up to a year: id, date, status and number of answers. An update asks all of the project's prompts to its AI models once. Use an id from here as the update input of get_overview, list_answers, get_sources or get_perception to read an earlier update; for trends across updates, use get_history instead. Read-only.

Set the status of one of the project's recommendations, as on the app's Recommendations page: "doing" when the work starts, "done" once it is applied (the next updates show whether the AI answers changed), "dismissed" when it does not apply, "todo" to reopen it. To mark a "source" or "content" recommendation done, give published_url: the page that now mentions the brand, or the page published (ask the user for it, never guess it); Rempa opens it to check it, then looks for it in the AI answers of the next updates. set_aside_days hides an open recommendation for 14 or 30 days (it comes back by itself), 0 brings it back now; give it instead of a status, or with one. The only tool that writes: it changes the status, the page published and the set-aside date, nothing else; it can be undone and repeating it is harmless. Returns the recommendation's id, title, status, page published and set-aside date. Paid plans only; an OAuth connection needs the permission to update.